Reallocations in teams of UAVs using dynamic programming and mixed initiative interactions

Abstract

We discuss the problem of dynamic reallocation of vehicles among teams of Unmanned Air Vehicles (UAV) executing concurrently. Each team addresses a task that consists of a sequence of subtasks to be executed in an adversary environment, where the vehicles face the risk of becoming inoperative. Our approach consists on separating the problem into a planning procedure followed by an optimal control problem, which is solved using stochastic dynamic programming (DP). We consider mixed-initiative interactions, where human operators are able to tune parameters of the problem according to their experience. The main goal of execution control is to balance the performance of teams in order to increase the success of the overall mission.